505-1

METHOD 505

ANALYSIS OF ORGANOHALIDE PESTICIDES AND COMMERCIAL POLYCHLORINATED BIPHENYL (PCB) PRODUCTS

IN WATER BY MICROEXTRACTION AND GAS CHROMATOGRAPHY

Revision 2.1

Edited by J.W. Munch (1995)

T. W. Winfield - Method 505, Revision 1.0 (1986)

T. W. Winfield - Method 505, Revision 2.0 (1989)

NATIONAL EXPOSURE RESEARCH LABORATORYOFFICE OF RESEARCH AND DEVELOPMENT

U.S. ENVIRONMENTAL PROTECTION AGENCYCINCINNATI, OHIO 45268

505-2

METHOD 505

ANALYSIS OF ORGANOHALIDE PESTICIDES AND COMMERCIALPOLYCHLORINATED BIPHENYL (PCB) PRODUCTS IN WATER BY

MICROEXTRACTION AND GAS CHROMATOGRAPHY

1.0 SCOPE AND APPLICATION

1.1 This method is applicable to the determination of the following analytes in1,2,3

finished drinking water, drinking water during intermediate stages of treatment,and the raw source water:

Analyte Registry NumberChemical Abstract Services

Alachlor 15972-60-8Aldrin 309-00-2Atrazine 1912-24-9Chlordane 57-74-9alpha-Chlorodane 5103-71-9gamma-Chlorodane 5103-74-2Dieldrin 60-57-1Endrin 72-20-8Heptachlor 76-44-8Heptachlor Epoxide 1024-57-3Hexachlorobenzene 118-74-1Hexachlorocyclopentadiene 77-74-4Lindane 58-89-9Methoxychlor 72-43-5cis-Nonachlor 5103-73-1trans-Nonachlor 39765-80-5Simazine 122-34-9Toxaphene 8001-35-2Aroclor 1016 12674-11-2Aroclor 1221 11104-28-2Aroclor 1232 11141-16-5Aroclor 1242 53469-21-9Aroclor 1248 12672-29-6Aroclor 1254 11097-69-1Aroclor 1260 11096-82-5

1.2 The analyst must demonstrate the applicability of the method by collectingprecision and accuracy data on fortified samples (i.e., groundwater, tap water)4

and provide qualitative confirmation of results by Gas Chromatography/MassSpectrometry (GC/MS) , or by GC analysis using dissimilar columns.5

505-3

1.3 Method detection limits (MDL) for the above organohalides and Aroclors have6

been experimentally determined (Section 13.2). Actual detection limits are highlydependent upon the characteristics of the gas chromatographic system used(e.g., column type, age, and proper conditioning; detector condition; and injectormode and condition).

1.4 This method is restricted to use by or under the supervision of analystsexperienced in the use of GC and in the interpretation of gas chromatograms.Each analyst must demonstrate the ability to generate acceptable results with thismethod using the procedure described in Section 11.0.

2.0 SUMMARY OF METHOD

2.1 Thirty-five mL of sample are extracted with 2 mL of hexane. One to 2 µL of theextract are then injected into a gas chromatograph equipped with a linearizedelectron capture detector for separation and analysis. Analytes are quantitatedusing procedural standard calibration (Section 3.12).

2.2 The extraction and analysis time is 40-70 minutes per sample depending upon theanalytes and the analytical conditions chosen. (See Section 6.9).

3.0 DEFINITIONS

3.1 Laboratory Duplicates (LD1 and LD2) -- Two sample aliquots taken in theanalytical laboratory and analyzed separately with identical procedures. Analysesof LD1 and LD2 give a measure of the precision associated with laboratoryprocedures, but not with sample collection, preservation, or storage procedures.

3.2 Field Duplicates (FD1 and FD2) -- Two separate samples collected at the sametime and place under identical circumstances and treated exactly the samethroughout field and laboratory procedures. Analyses of FD1 and FD2 give ameasure of the precision associated with sample collection, preservation andstorage, as well as with laboratory procedures.

3.3 Laboratory Reagent Blank (LRB) -- An aliquot of reagent water that is treatedexactly as a sample including exposure to all glassware, equipment, solvents,reagents, internal standards, and surrogates that are used with other samples.The LRB is used to determine if method analytes or other interferences arepresent in the laboratory environment, the reagents, or the apparatus.

3.4 Field Reagent Blank (FRB) -- Reagent water placed in a sample container in thelaboratory and treated as a sample in all respects, including exposure to samplingsite conditions, storage, preservation and all analytical procedures. The purposeof the FRB is to determine if method analytes or other interferences are presentin the field environment.

505-4

3.5 Laboratory Performance Check Solution (LPC) -- A solution of method analytes,surrogate compounds, and internal standards used to evaluate the performanceof the instrument system with respect to a defined set of method criteria.

3.6 Laboratory Fortified Blank (LFB) -- An aliquot of reagent water to which knownquantities of the method analytes are added in the laboratory. The LFB isanalyzed exactly like a sample, and its purpose is to determine whether themethodology is in control, and whether the laboratory is capable of makingaccurate and precise measurements at the required method detection limit.

3.7 Laboratory Fortified Sample Matrix (LFM) -- An aliquot of an environmentalsample to which known quantities of the method analytes are added in thelaboratory. The LFM is analyzed exactly like a sample, and its purpose is todetermine whether the sample matrix contributes bias to the analytical results.The background concentrations of the analytes in the sample matrix must bedetermined in a separate aliquot and the measured values in the LFM correctedfor background concentrations.

3.8 Stock Standard Solution -- A concentrated solution containing a single certifiedstandard that is a method analyte, or a concentrated solution of a single analyteprepared in the laboratory with an assayed reference compound. Stock standardsolutions are used to prepare primary dilution standards.

3.9 Primary Dilution Standard Solution (PDS) -- A solution of several analytesprepared in the laboratory from stock standard solutions and diluted as neededto prepare calibration solutions and other needed analyte solutions.

3.10 Calibration Standard (CAL) -- A solution prepared from the primary dilutionstandard solution and stock standard solutions of the internal standards andsurrogate analytes. The CAL solutions are used to calibrate the instrumentresponse with respect to analyte concentration.

3.11 Quality Control Sample (QCS) -- A sample matrix containing method analytes ora solution of method analytes in a water miscible solvent which is used to fortifyreagent water or environmental samples. The QCS is obtained from a sourceexternal to the laboratory, and is used to check laboratory performance withexternally prepared test materials.

3.12 Procedural Standard Calibration -- A calibration method where aqueouscalibration standards are prepared and processed (e.g., purged, extracted, and/orderivatized) in exactly the same manner as a sample. All steps in the processfrom addition of sampling preservatives through instrumental analyses areincluded in the calibration. Using procedural standard calibration compensatesfor any inefficiencies in the processing procedure.

505-5

4.0 INTERFERENCES

4.1 Method interferences may be caused by contaminants in solvents, reagents,glassware and other sample processing apparatus that lead to discrete artifacts orelevated baselines in gas chromatograms. All reagents and apparatus must beroutinely demonstrated to be free from interferences under the conditions of theanalysis by running laboratory reagent blanks as described in Section 9.2.

4.1.1 Glassware must be scrupulously cleaned . Clean all glass- ware as soon2

as possible after use by thoroughly rinsing with the last solvent used init. Follow by washing with hot water and detergent and thorough rinsingwith tap and reagent water. Drain dry, and heat in an oven or mufflefurnace at 400°C for one hour. Do not heat volumetric ware. Thermallystable materials, such as PCBs, might not be eliminated by this treatment.Thorough rinsing with acetone may be substituted for the heating. Afterdrying and cooling, seal and store glassware in a clean environment toprevent any accumulation of dust or other contaminants. Store invertedor capped with aluminum foil.

4.1.2 The use of high purity reagents and solvents helps to minimizeinterference problems. Purification of solvents by distilla- tion in all-glasssystems may be required.

Warning: When a solvent is purified, stabilizers put into the solvent bythe manufacturer are removed thus potentially making the solventhazardous. Also, when a solvent is purified, preservatives put into thesolvent by the manufacturer are removed thus potentially reducing theshelf-life.

4.2 Interfering contamination may occur when a sample containing lowconcentrations of analytes is analyzed immediately following a sample containingrelatively high concentrations of analytes. Between-sample rinsing of the samplesyringe and associated equipment with hexane can minimize sample crosscontamination. After analysis of a sample containing high concentrations ofanalytes, one or more injections of hexane should be made to ensure that accuratevalues are obtained for the next sample.

4.3 Matrix interferences may be caused by contaminants that are coextract- ed fromthe sample. Also, note that all the analytes listed in the scope and applicationsection are not resolved from each other on any one column, i.e., one analyte ofinterest may be an interferant for another analyte of interest. The extent of matrixinterferences will vary considerably from source to source, depending upon thewater sampled. Cleanup of sample extracts may be necessary. Analyteidentifications should be confirmed (Section 11.4).

505-6

4.4 It is important that samples and working standards be contained in the samesolvent. The solvent for working standards must be the same as the final solventused in sample preparation. If this is not the case, chromatographic comparabilityof standards to sample may be affected.

4.5 Caution must be taken in the determination of endrin since it has been reportedthat the splitless injector may cause endrin degradation . The analyst should be7

alerted to this possible interference resulting in an erratic response for endrin.

4.6 Variable amounts of pesticides and commercial PCB products from aqueoussolutions adhere to glass surfaces. It is recommended that sample transfers andglass surface contacts be minimized, and that adequate rinsing of glass surfacesbe performed.

4.7 Aldrin, hexachlorocyclopentadiene, and methoxychlor are rapidly oxidized bychlorine. Dechlorination with sodium thiosulfate at time of collection will stopfurther oxidation of these compounds.

Warning: An interfering, erratic peak has been observed within the retentionwindow of heptachlor during many analyses of reagent, tap, and groundwater.It appears to be related to dibutyl phthalate; however, the specific source has notyet been definitively determined. The observed magnitude and character of thispeak randomly varies in numerical value from successive injections made fromthe same vial.

5.0 SAFETY

5.1 The toxicity and carcinogenicity of chemicals used in this method have not beenprecisely defined; each chemical should be treated as a potential health hazard,and exposure to these chemicals should be minimized. Each laboratory isresponsible for maintaining awareness of OSHA regulations regarding safehandling of chemicals used in this method. Additional references to laboratorysafety are available for the information of the analyst.8-10

5.2 The following organohalides have been tentatively classified as known orsuspected human or mammalian carcinogens: aldrin, commercial PCB products,chlordane, dieldrin, heptachlor, hexachlorobenzene, and toxaphene. Purestandard materials and stock standard solutions of these compounds should behandled in a hood or glovebox.

Warning: When a solvent is purified, stabilizers put into the solvent by themanufacturer are removed thus potentially making the solvent hazardous.

505-7

6.0 EQUIPMENT AND SUPPLIES (All specifications are suggested. Catalog numbers areincluded for illustration only.)

6.1 Sample Containers -- 40 mL screw cap vials (Pierce #13075 or equivalent) eachequipped with a size 24 cap with a flat, disc-like TFE facing backed with apolyethylene film/foam extrusion (Fisher #02-883-3F or equivalent). Prior to use,wash vials and septa with detergent and rinse with tap and distilled water.Allow the vials and septa to air dry at room temperature, place the vials in a400°C oven for one hour, then remove and allow to cool in an area known to befree of organics.

6.2 Vials -- auto sampler, screw cap with septa, 1.8 mL, Varian #96-000099-00 orequivalent or any other autosampler vials not requiring more than 1.8 mL samplevolumes.

6.3 Auto Sampler -- Hewlett-Packard 7671A, or equivalent.

6.4 Micro Syringes -- 10 µL and 100 µL.

6.5 Micro Syringe -- 25 µL with a 2 inch by 0.006 inch needle - Hamilton 702N orequivalent.

6.6 Pipettes -- 2.0 mL and 5.0 mL transfer.

6.7 Volumetric Flasks -- 10 mL and 100 mL, glass stoppered.

6.8 Standard Solution Storage Containers -- 15 mL bottles with PTFE-lined screwcaps.

6.9 Gas Chromatograph -- Analytical system complete with temperatureprogrammable GC and split/splitless injector suitable for use with capillarycolumns and all required accessories including syringes, analytical columns, gases,a linearized electron capture detector and stripchart recorder. A data system isrecommended for measuring peak areas. Table 1 lists retention times observedfor method analytes using the columns and analytical conditions described below.

6.9.1 Three gas chromatographic columns are recommended. Column 1(Section 6.9.2) should be used as the primary analytical column unlessroutinely occurring analytes are not adequately resolved. Validation datapresented in this method were obtained using this column. Columns 2and 3 are recommended for use as confirmatory columns when GC/MSconfirmation is not available. Alternative columns may be used inaccordance with the provisions described in Section 9.4.

6.9.2 Column 1 (primary column) - 0.32 mm ID x 30 M long fused silicacapillary with chemically bonded methyl polysiloxane phase (DB-1, 1.0 µmfilm, or equivalent). Helium carrier gas flow is about 25 cm/sec. linearvelocity, measured at 180° with 9 psi column head pressure. The oven

505-8

temperature is programmed from 180-260°C at 4°C/min. and held at260°C until all expected compounds have eluted. Injector temperature:200°C. Splitless Mode: 0.5 minute. Detector temperature: 290°C. Samplechromatograms for selected pesticides are presented in Figures 1 and 2.Chromatograms of the Aroclors, toxaphene, and technical chlordane arepresented in Figures 3 through 11.

6.9.3 Column 2 (alternative Column 1) - 0.32mm ID x 30 M long fused silicacapillary with a 1:1 mixed phase of dimethyl silicone and polyethyleneglycol (Durawax-DX3, 0.25 µm film, or equivalent). Helium carrier gasflow is about 25 cm/sec. linear velocity and oven temperature isprogrammed from 100-210°C at 8°C/min., and held at 210°C until allexpected compounds have eluted. Then the post temperature isprogrammed to 240°C at 8°C/min. for five minutes.

6.9.4 Column 3 (alternative Column 2) - 0.32mm ID x 25 M long fused silicacapillary with chemically bonded 50:50 Methyl-Phenyl silicone (OV-17,1.5 µm film thickness, or equivalent). Helium carrier gas flow is about40 cm/sec. linear velocity and oven temperature is programmed from100-260°C at 4°C/min. and held at 260°C until all expected compoundshave eluted.

7.0 REAGENTS AND STANDARDS

Warning: When a solvent is purified, stabilizers put into the solvent by themanufacturer are removed thus potentially making the solvent hazardous. Also, whena solvent is purified, preservatives put into the solvent by the manufacturer are removedthus potentially making the shelf-life short.

7.1 Reagents

7.1.1 Hexane Extraction Solvent -- UV Grade, Burdick and Jackson #216 orequivalent.

7.1.2 Methyl Alcohol, ACS Reagent Grade -- Demonstrated to be free ofanalytes.

7.1.3 Sodium Chloride, NaCl, ACS Reagent Grade -- For pretreatment beforeuse, pulverize a batch of NaCl and place in a muffle furnace at roomtemperature. Increase the temperature to 400°C and hold for 30 minutes.Store in a glass (not plastic) bottle to avoid phthalate contamination..

7.1.4 Sodium thiosulfate, Na S O , ACS Reagent Grade -- For preparation of2 2 3

solution (0.04 g/mL), mix 1 g of Na S O with reagent water and bring to2 2 3

25 mL volume in a volumetric flask. Verify the stability of this solutionand replace as necessary.

505-9

7.2 Reagent Water -- Reagent water is defined as water free of interference whenemployed in the procedure described herein.

7.2.1 A Millipore Super-Q Water System or its equivalent may be used togenerate deionized reagent water.

7.2.2 Test reagent water each day it is used by analyzing it according toSection 9.2.

7.3 Stock Standard Solutions (SSS) -- These solutions may be obtained as certifiedsolutions or prepared from pure standard materials using the followingprocedures:

7.3.1 Prepare stock standard solutions (5000 µg/mL) by accurately weighingabout 0.0500 g of pure material. Dissolve the material in methanol anddilute to volume in a 10 mL volumetric flask. Larger volumes can be usedat the convenience of the analyst. When compound purity is assayed tobe 96% or greater, the weight can be used without correction to calculatethe concentration of the stock standard. Commercially prepared stockstandards can be used at any concentration if they are certified by themanufacturer or by an independent source.

7.3.2 Transfer the stock standard solutions into Teflon-sealed screw-cap bottles.Store at 4°C and protect from light. Stock standard solutions should bechecked frequently for signs of degradation or evaporation, especially justprior to preparing calibration standards from them.

7.3.3 Stock standard solutions must be replaced after six months, or sooner ifcomparison with check standards indicates a problem.

7.4 Primary Dilution Standard Solutions (PDS) -- Use stock standard solutions toprepare primary dilution standard solutions that contain the analytes in methanol.The primary dilution standards should be prepared at concentrations that can beeasily diluted to prepare aqueous calibration standards (Section 10.2.1) that willbracket the working concentra- tion range. Store the primary dilution standardsolutions with minimal headspace and check frequently for signs of deteriorationor evaporation, especially just before preparing calibration standards. The storagetime described for stock standard solutions in Section 7.3.3 also applies to primarydilution standard solutions.

Note: Primary dilution standards for toxaphene, chlordane and each of theAroclors must be prepared individually.

505-10

8.0 SAMPLE COLLECTION, PRESERVATION, AND STORAGE

8.1 Sample Collection

8.1.1 Collect all samples in 40 mL bottles into which 3 mg of sodium thiosulfatecrystals have been added to the empty bottles just prior to shipping to thesampling site. Alternately, 75 µL of freshly prepared sodium thiosulfatesolution (0.04 g/mL) may be added to empty 40 mL bottles just prior tosample collection.

8.1.2 When sampling from a water tap, open the tap and allow the system toflush until the water temperature has stabilized (usually about10 minutes). Adjust the flow to about 500 mL/min. and collect samplesfrom the flowing stream.

8.1.3 When sampling from a well, fill a wide-mouth bottle or beaker withsample, and carefully fill 40 mL sample bottles.

8.2 Sample Preservation

8.2.1 The samples must be chilled to 4°C at the time of collection andmaintained at that temperature until the analyst is prepared for theextraction process. Field samples that will not be received at thelaboratory on the day of collection must be packaged for shipment withsufficient ice to insure that they will be maintained at 4°C until arrival atthe laboratory.

8.3 Sample Storage

8.3.1 Store samples and extracts at 4°C until extraction and analysis.

8.3.2 Extract all samples as soon as possible after collection. Results of holdingtime studies suggest that all analytes with the possible exception ofheptachlor were adequately stable for 14 days when stored under theseconditions. In general, heptachlor showed inconsistent results. Ifheptachlor is to be determined, samples should be extracted within sevendays of collection. The maximum holding time for all other analytes is14 days. Analyte stability may be affected by the matrix; therefore, theanalyst should verify that the preservation technique is applicable to thesamples under study.

8.3.3 Because of the potential for extract volume loss due to evaporation, it isrecommended that extracts be analyzed immediately after preparation. Ifthis is not possible, extracts may be stored at 4°C or less in tightly cappedvials (Section 6.2) for up to 24 hours.

505-11

9.0 QUALITY CONTROL

9.1 Minimum quality control (QC) requirements are initial demonstration oflaboratory capability, analysis of laboratory reagent blanks (LRB), laboratoryfortified blanks (LFB), laboratory fortified sample matrix (LFM), and qualitycontrol samples (QCS). A MDL for each analyte must also be determined.

9.2 Laboratory Reagent Blanks (LRB) -- Before processing any samples, the analystmust demonstrate that all glassware and reagent interferences are under control.Each time a set of samples is extracted or reagents are changed, an LRB must beanalyzed. If within the retention time window of any analyte the LRB producesa peak that would prevent the determination of that analyte, determine the sourceof contamination and eliminate the interference before processing samples.

9.3 Initial Demonstration of Capability

9.3.1 Select a representative fortified concentration (about 10 times MDL or ata concentration in the middle of the calibration range established inSection 10.0) for each analyte. Prepare a standard concentrate containingeach analyte at 1000 times the selected concentration. With a syringe, add35 µL of the concentrate to each of four to seven 35 mL aliquots of reagentwater, and analyze each aliquot according to procedures beginning inSection 11.0.

9.3.2 For each analyte the mean recovery value for these samples must fall inthe range of ±30% of the fortified amount. The RSD for thesemeasurements must be 20% or less. For those compounds that meet theacceptance criteria, performance is considered acceptable. For thosecompounds that fail these criteria, this procedure must be repeated usingfresh replicate samples until satisfactory performance has beendemonstrated.

9.3.3 For each analyte, determine the MDL. Prepare a minimum of seven LFBsat a low concentration. Use calibration data obtained in Section 10.0 toestimate a concentration for each analyte that will produce a peak with athree to five times signal to noise response. Extract and analyze eachreplicate according to Sections 11.0 and 12.0. It is recommended that theseLFBs be prepared and analyzed over a period of several days, so that dayto day precision is reflected in the precision measurements. Calculatemean recovery and standard deviation for each analyte. Use the equationgiven in Section 13.0 to calculate the MDL.

9.3.4 The initial demonstration of capability is used primarily to preclude alaboratory from analyzing unknown samples via a new, unfamiliarmethod prior to obtaining some experience with it. It is expected that aslaboratory personnel gain experience with this method the quality of datawill improve beyond those required here.

505-12

9.4 The analyst is permitted to modify GC columns, GC conditions, concentrationtechniques (i.e., evaporation techniques), internal standards or surrogatecompounds. Each time such method modifications are made, the analyst mustrepeat the procedures in Section 9.3.

9.5 Assessing Laboratory Performance -- Laboratory Fortified Blank (LFB)

9.5.1 The laboratory must analyze at least one LFB per sample set (all samplesextracted within a 24-hour period). If the sample set contains more than20 samples, analyze one LFB for every 20 samples. Ideally the fortifyingconcentration of each analyte in the LFB sample should be the same asthat selected in Section 9.3.1. Calculate accuracy as percent recovery (X ).i

If the recovery of any analyte falls outside the control limits (seeSection 9.5.2), that analyte is judged out of control, and the source of theproblem should be identified and resolved before continuing analyses.Because this LFB is prepared and analyzed in the same way as acalibration verification standard, it can be used to satisfy the calibrationrequirement in Section 10.2.3.

Note: It is suggested that one multi-component analyte (toxaphene,chlordane or an Aroclor) LFB also be analyzed with each sample set. Byselecting a different multi-component analyte for this LFB each work shift,LFB data can be obtained for all of these analytes over the course ofseveral days

9.5.2 Until sufficient data become available from within their own laboratory,usually a minimum of results from 20-30 analyses, the laboratory mayassess laboratory performance against the control limits in Section 9.3.2that are derived from the data in Table 2. When sufficient internalperformance data becomes available, develop control limits from the meanpercent recovery ( ) and standard deviation (S) of the percent recovery.These data are used to establish upper and lower control limits as follows:

UPPER CONTROL LIMIT = + 3SLOWER CONTROL LIMIT = - 3S

After each five to 10 new recovery measurements, new control limitsshould be calculated using only the most recent 20-30 data points. Thesecalculated control limits should not exceed the fixed limits in Section 9.3.2.

505-13

9.5.3 It is recommended that the laboratory periodically determine anddocument its detection limit capabilities for analytes of interest.

Caution: No attempts to establish low detection limits should be madebefore instrument optimization and adequate conditioning of both thecolumn and the GC system. Conditioning includes the processing of LFBand LFM samples containing moderate concentration levels of theseanalytes.

9.5.4 At least quarterly the laboratory should analyze quality control samples(QCS). If acceptance criteria are not met, corrective action should be takenand documented.

9.6 Assessing Analyte Recovery -- Laboratory Fortified Sample Matrix (LFM)

9.6.1 The laboratory must add a known concentration to a minimum of 10% ofthe routine samples or one LFM per set, whichever is greater. Thefortified concentration should not be less than the backgroundconcentration of the sample selected for fortification. Ideally the LFMconcentration should be the same as that used for the LFB (Section 9.3.1).Periodically, samples from all routine sample sources should be fortified.

9.6.2 Calculate the percent recovery (R ) for each analyte, corrected fori

background concentrations measured in the unfortified sample.

9.6.3 If the recovery of any such analyte falls outside the range of ±35% of thefortified amount, and the laboratory performance for that analyte is shownto be in control (Section 9.5), the recovery problem encountered with thedosed sample is judged to be matrix related, not system related. Theresult for that analyte in the unfortified sample is labeled suspect/matrixto inform the data user that the results are suspect due to matrix effects.

9.7 The laboratory may adopt additional quality control practices for use with thismethod. The specific practices that are most productive depend upon the needsof the laboratory and the nature of the samples. For example, field or laboratoryduplicates may be analyzed to assess the precision of the environmentalmeasurements or field reagent blanks may be used to assess contamination ofsamples under site conditions, transportation and storage.

10.0 CALIBRATION AND STANDARDIZATION

10.1 Establish GC operating parameters equivalent to those indicated in Section 6.9.

Warning: Endrin is easily degraded in the injection port if the injection port orfront of the column is dirty. This is the result of buildup of high boiling residuefrom sample injection. Check for degradation problems daily by injecting a mid-level standard containing only endrin. Look for the degradation products ofendrin (endrin ketone and endrin aldehyde). If degradation of endrin exceeds

505-14

20%, take corrective action before proceeding with calibration. Calculate percentbreakdown as follows:

10.2 At least three calibration standards are needed; five are recommended. Guidanceon the number of standards is as follows: A minimum of three calibrationstandards are required to calibrate a range of a factor of 20 in concentration. Fora factor of 50 use at least four standards, and for a factor of 100 at least fivestandards. The lowest standard should represent analyte concentrations near, butabove, their respective MDLs. The remaining standards should bracket theanalyte concentrations expected in the sample extracts, or should define theworking range of the detector.

10.2.1 To prepare a calibration standard (CAL), add an appropriate volume ofa secondary dilution standard to a 35 mL aliquot of reagent water in a40 mL bottle. Do not add less than 20 µL of an alcoholic standard to thereagent water. Use a 25 µL micro syringe and rapidly inject themethanolic standard into the middle point of the water volume. Removethe needle as quickly as possible after injection. Mix by inverting andshaking the capped bottle several times. Aqueous standards must beprepared fresh daily.

10.2.2 Starting with the standard of lowest concentration, prepare, extract, andanalyze each calibration standard beginning with Section 11.2 and tabulatepeak height or area response versus the concentration in the standard.The results are to be used to prepare a calibration curve for eachcompound by plotting the peak height or area response versus theconcentration. Alternatively, if the ratio of concentration to response(calibration factor) is a constant over the working range (20% RSD or less),linearity to the origin can be assumed and the average ratio or calibrationfactor can be used in place of a calibration curve.

Note: Toxaphene, chlordane, and Aroclor standards must be injectedseparately. See Section 12.2 for information on quantitation of multi-component analytes.

10.2.3 The working calibration curve or calibration factor must be verified oneach working day by the measurement of a minimum of two calibrationcheck standards, one at the beginning and one at the end of the analysisday. These check standards should be at two different concentrationlevels to verify the calibration curve. For extended periods of analysis(greater than eight hours), it is strongly recommended that checkstandards be interspersed with samples at regular intervals during thecourse of the analyses. If the response for any analyte varies from the

505-15

predicted response by more than the criteria in Section 9.3.2, the test mustbe repeated using a fresh calibration standard. If the results still do notagree, generate a new calibration curve. For those analytes that failed thecalibration verification, results from field samples analyzed since the lastpassing calibration should be considered suspect. Reanalyze sampleextracts for these analytes after acceptable calibration is restored.

Note: It is suggested that a calibration verification standard of one multi-component analyte, either chlordane, toxaphene or an Aroclor also beanalyzed each work shift. By selecting a different multi-componentanalyte for this calibration verification each work shift, continuingcalibration data can be obtained for all of these analytes over the courseof several days.

10.3 Instrument Performance -- Check the performance of the entire analytical systemdaily using data gathered from analyses of laboratory reagent blanks (LRB),(CAL), and laboratory duplicate samples (LD1 and LD2).

10.3.1 Significant peak tailing in excess of that shown for the target compoundsin the method chromatograms (Figures 1-11) must be corrected. Tailingproblems are generally traceable to active sites on the GC column,improper column installation, or operation of the detector.

10.3.2 Check the precision between replicate analyses. Poor precision isgenerally traceable to pneumatic leaks, especially at the injection port. Ifthe GC system is apparently performing acceptably but with decreasedsensitivity, it may be necessary to generate a new curve or set ofcalibration factors to verify the decreased responses before searching forthe source of the problem.

10.3.3 Observed relative area responses of endrin (Section 10.1) must meet thefollowing general criteria:

10.3.3.1 The breakdown of endrin into its aldo and keto forms mustbe adequately consistent during a period in which a seriesof analyses is made. Equivalent relative amounts ofbreakdown should be demonstrated in the LRB, LFB, CAL,and QCS. Consistent break- down resulting in theseanalyses would suggest that the breakdown occurred in theinstrument system and that the methodology is in control.

10.3.3.2 Analyses of laboratory fortified matrix (LFM) samples mustalso be adequately consistent after corrections for potentialbackground concentrations are made.

505-16

11.0 PROCEDURE

11.1 Sample Preparation

11.1.1 Remove samples from storage and allow them to equilibrate to roomtemperature.

11.1.2 Remove the container caps. Withdraw and discard a 5 mL volume usinga 10 mL graduated cylinder. Replace the container caps and weigh thecontainers with contents to the nearest 0.1 g and record these weights forsubsequent sample volume determinations (Section 11.3).

11.2 Extraction and Analysis

11.2.1 Remove the container cap of each sample, and add 6 g NaCl(Section 7.1.3) to the sample bottle. Using a class A, TD transfer orautomatic dispensing pipet, add 2.0 mL of hexane. Recap and shakevigorously by hand for one minute. Invert the bottle and allow the waterand hexane phases to separate.

11.2.2 Remove the cap and carefully transfer approximately 0.5 mL of hexanelayer into an autosampler vial using a disposable glass pipet.

11.2.3 Transfer the remaining hexane phase, being careful not to include any ofthe water phase, into a second autosampler vial. Reserve this second vialat 4°C for reanalysis if necessary.

11.2.4 Transfer the first sample vial to an autosampler set up to inject 1-2 µLportions into the gas chromatograph for analysis (See Section 6.9 for GCconditions). Alternately, 1-2 µL portions of samples, blanks, andstandards may be manually injected, although an autosampler is stronglyrecommended.

11.3 Determination of Sample Volume in Bottles Not Calibrated

11.3.1 Discard the remaining sample/hexane mixture from the sample bottle.Shake off the remaining few drops using short, brisk wrist movements.

11.3.2 Reweigh the empty container with original cap and calculate the netweight of sample by difference to the nearest 0.1 g (Section 11.1.2 throughSection 11.3.2). This net weight (in grams) is equivalent to the volume(in mL) of water extracted (Section 12.4). Alternatively, by using 40 mLbottles precalibrated at 35 mL levels, the gravimetric steps can be omitted,thus increasing the speed and ease of this extraction process.

505-17

11.4 Identification of Analytes

11.4.1 Identify a sample component by comparison of its retention time to theretention time of a reference chromatogram. If the retention time of anunknown compound corresponds, within limits, to the retention time ofa standard compound, then identification is considered positive.

11.4.2 The width of the retention time window used to make identifi- cationsshould be based upon measurements of actual retention time variationsof standards over the course of a day. Three times the standard deviationof a retention time can be used to calculate a suggested window size fora compound. However, the experience of the analyst should weighheavily in the interpretation of chromatograms.

11.4.3 Identification requires expert judgement when sample compo- nents arenot resolved chromatographically. When peaks obviously represent morethan one sample component (i.e., broadened peak with shoulder(s) orvalley between two or more maxima), or any time doubt exists over theidentification of a peak on a chromatogram, appropriate alternativetechniques to help confirm peak identification need be employed. Forexample, more positive identification may be made by the use of analternative detector which operates on a chemical/physical principledifferent from that originally used, e.g., mass spectrometry, or the use ofa second chromatography column. Suggested alternative columns aredescribed in Section 6.9.

Note: Identify multi-component analytes by comparison of the samplechromatogram to the corresponding calibration standard chromatogramsof chlordane, toxaphene and the Aroclors. Identification of multi-component analytes is made by pattern recognition, in which theexperience of the analyst is an important factor.

12.0 DATA ANALYSIS AND CALCULATIONS

12.1 Identify the organohalides in the sample chromatogram by comparing theretention time of the suspect peak to retention times generated by the calibrationstandards and the laboratory fortified blanks. Identify the multicomponentcompounds using all peaks that are characteristic of the specific compound fromchromatograms generated with individual standards.

12.2 To quantitate multi-component analytes, one of the following methods should beused.

Option 1 - Calculate an average response factor or linear regression equation foreach multi-component analyte using the combined area of all the componentpeaks in each of the calibration standard chromatograms.

505-18

Option 2 - Calculate an average response factor or linear regression equation foreach multi-component analyte using the combined areas of three to six of themost intense and reproducible peaks in each of the calibration standardchromatograms.

When quantifying multi-component analytes in samples, the analyst should usecaution to include only those peaks from the sample that are attributable to themulti-component analyte. Option 1 should not be used if there are significantinterference peaks within the chlordane, Aroclor or toxaphene pattern.

12.3 Use the multi-point calibration curve or calibration factor (Section 10.2.3) todirectly calculate the uncorrected concentration (Ci) of each analyte in the sample(e.g., calibration factor x response). Do not use the daily calibration standard toquantitate method analytes in samples. If any analyte response exceeds thecalibration range, dilute the extract and reanalyze.

12.4 Calculate the sample volume (V ) as equal to the net sample weight: s

V = gross weight (Section 11.1.2) - bottle tare (Section 11.3.2).s

12.5 Calculate the corrected sample concentration as:

12.6 Results should be reported with an appropriate number of significant figures.Experience indicates that three significant figures may be used for concentrationsabove 99 µg/L, two significant figures for concentrations between 1-99 µg/L, andone significant figure for lower concentrations.

13.0 METHOD PERFORMANCE

13.1 Single laboratory (NERL-Cincinnati) accuracy and precision at severalconcentrations in reagent, ground, and tap water matrices are presented inTable 2. These results were obtained from data generated with a DB-1 column,and with quantitation Option 2 as described in Section 12.2.

13.2 With these data, the method detection limits (MDL) in Table 2 were calculatedusing the formula:

505-19

where: t = Student's t value for the 99% confidence level with n-1 degrees(n-1,1-alpha = 0.99)

of freedom n = number of replicates S = standard deviation of replicate analyses.

13.3 This method has been tested by 10 laboratories using reagent water and

groundwater fortified at three concentration levels. Single operator precision,overall precision, and method accuracy were found to be directly related to theconcentration of the analyte and virtually independent of the sample matrix.Linear equations to describe the relationships are presented in Table 3.

14.0 POLLUTION PREVENTION

14.1 This method utilizes a microextraction procedure that requires the use of verysmall volumes of hexane, thus making this method safe for use by the laboratoryanalyst and harmless to the environment. For information concerning pollutionprevention that may be applicable to laboratory operations, consult "Less is Better:Laboratory Chemical Management for Waste Reduction" available from theAmerican Chemical Society's Department of Government Relations, and SciencePolicy, 1155 16th Street N.W., Washington, D.C. 20036.

15.0 WASTE MANAGEMENT

15.1 It is the laboratory's responsibility to comply with all federal, state, and localregulations governing the waste management, particularly the hazardous wasteidentification rules and land disposal restrictions, and to protect the air, water,and land by minimizing and controlling all releases from fume hoods and benchoperations. Also, compliance is required with any sewage discharge permits andregulations. For further information on waste management, consult "The WasteManagement Manual for Laboratory Personnel," also available from the AmericanChemical Society at the address in Section 14.1.

16.0 REFERENCES

1. Glaze, W.W., Lin, C.C. Optimization of Liquid-Liquid Extraction Methods forAnalysis of Organics in Water, EPA-600/S4-83-052, January 1984.

2. Henderson, J.E., Peyton, G.R. and Glaze, W.H. (1976). In "Identification and Analysisof Organic Pollutants in Water" (L.H. Keith ed.), pp. 105-111. Ann Arbor Sci. Publ.,Ann Arbor, Michigan.

505-20

3. Richard, J.J., Junk, G.A. "Liquid Extraction for Rapid Determination of Halomethanesin Water," Journal AWWA, 69, 62, January 1977.

4. "Handbook for Analytical Quality Control in Water and Wastewater Laboratories",EPA-600/4-79-019, U. S. Environmental Protection Agency, Environmental Monitoringand Support Laboratory, Cincinnati, Ohio 45268, March 1979.

5. Munch, J. W. "Method 525.2-Determination of Organic Compounds in DrinkingWater by Liquid-Solid Extraction and Capillary Column Chromatography/MassSpectrometry" in Methods for the Determination of Organic Compounds in DrinkingWater; Supplement 3 (1995). USEPA, National Exposure Research Laboratory,Cincinnati, Ohio 45268.

6. Glaser, J.A. et al. "Trace Analyses for Wastewaters", Environmental Science andTechnology, 15, 1426 (1981).

7. Bellar, T.A., Stemmer, P., Lichtenberg, J.J. "Evaluation of Capillary Systems for theAnalysis of Environmental Extracts", EPA-600/S4-84-004, March 1984.

8. "Carcinogens-Working with Carcinogens", Department of Health, Education, andWelfare, Public Health Service, Center for Disease Control, National Institute ofOccupational Safety and Health, Publication No. 77-206, August 1977.

9. "OSHA Safety and Health Standards, General Industry", (29CFR1910), OccupationalSafety and Health Administration, OSHA 2206, (Revised, January 1976).

10. "Safety in Academic Chemistry Laboratories", American Chemical Society Publication,Committee on Chemical Safety, 3rd Edition, 1979.

505-21

17.0 TABLES, DIAGRAMS, FLOWCHARTS, AND VALIDATION DATA

TABLE 1. RETENTION TIMES FOR METHOD ANALYTES

Retention Time , Mina

Analyte Primary Confirm. 1 Confirm. 2

Hexachlorocyclopentadiene 5.5 6.8 5.2Simazine 10.9 25.7 19.9Atrazine 11.2 22.6 19.6Hexachlorobenzene 11.9 13.4 15.6Lindane 12.3 18.4 18.7Alachlor 15.1 19.7 21.1Heptachlor 15.9 17.5 20.0Aldrin 17.6 18.4 21.4Heptachlor Epoxide 19.0 24.6 24.6gamma-Chlordane 19.9 25.9 26.0alpha-Chlordane 20.9 26.6 26.6trans-Nonachlor 21.3 24.8 26.3Dieldrin 22.1 45.1 27.8Endrin 23.2 33.3 29.2cis-Nonachlor 24.3 39.0 30.4Methoxychlor 30.0 58.5 36.4

Primaryb

Aroclor 1016 13.6, 14.8, 15.2, 16.2, 17.7Aroclor 1221 7.7, 9.0, 15.9, 19.1, 24.7Aroclor 1232 11.2, 14.7, 13.6, 15.2, 17.7Aroclor 1242 11.2, 13.6, 14.7, 15.2, 17.7, 19.8Aroclor 1248 14.8, 16.2, 17.1, 17.7, 19.8, 22.0Aroclor 1254 19.1, 21.9, 23.4, 24.9, 26.7Aroclor 1260 23.4, 24.9, 26.7, 28.2, 29.9, 21.3Chlordane 15.1, 15.9, 20.1, 20.9, 21.3Toxaphene 21.7, 22.5, 26.7, 27.2

Columns and analytical conditions are described in Sections 6.9.2 through 6.9.4.a

Column and conditions described in Section 6.9.2. More than one peak listed doesb

not indicate the total number of peaks characteristic of the multi-componentanalyte. Listed peaks indicate only the ones chosen for summation in thequantification.

505-22

TABLE 2. SINGLE LABORATORY ACCURACY, PRECISION, AND METHOD DETECTION LIMITS (MDLs) FORANALYTES FROM REAGENT WATER, GROUND WATER, AND TAP WATERa

Accuracy and Standard Deviation Data

AnalyteMDLµg/L

Concen-tration*

µg/L

Reagent Water Ground Water Tap Water

Rc SRd R SR R SR

AlachlorAldrin

Atrazine

alpha-Chlordane

gamma-Chlordane

Chlordane

Dieldrin

Endrin

Heptachlor

Heptachlor Epoxide

Hexachlorobenzene

Hexachlorocyclopentadiene

Lindane

Methoxychlor

0.2250.0072.4

0.006

0.012

0.14

0.012

0.063

0.003

0.004

0.002

0.13

0.003

0.96

0.500.055.0

20.00.060.350.060.350.173.40.103.60.103.60.0321.20.041.40.0030.090.150.350.031.22.107.03

102 106

85 95 95 86 95 86

– –

87 114 119

99 77 80

100 115 104 103

73 73 91

111 100

98

13.4 20.0 16.2

5.2 3.5

17.0 0.4

18.5 8.0 3.6

17.1 9.1

29.8 6.5

10.2 7.4

15.6 6.6

13.5 6.6 5.1

11.7 6.5 5.0

21.0 10.9

– 86 95 86 83 94 86 95

– –

67 94 94

100 37 71 90

103 91

101 87 69 88

109 – –

– 16.3

7.3 9.1 4.4

10.2 5.3

14.5 – –

10.1 8.6

20.2 11.3

6.8 9.8

14.2 6.9

10.9 4.4 5.1 4.8 7.7 3.4

– –

– –

108 91 85 91 83 91

105 95 92 81

106 85

200 106 112

81 100

88 191 109 103

93 – –

– –

10.9 3.1 7.1 2.4

14.7 6.0

12.4 9.6

15.7 14.0 14.0 12.4 22.6 16.8

7.5 5.9

15.6 13.4 18.5 14.3

8.1 18.4

– –

505-23

TABLE 2. SINGLE LABORATORY ACCURACY, PRECISION, AND METHOD DETECTION LIMITS (MDLs) FORANALYTES FROM REAGENT WATER, GROUND WATER, AND TAP WATERa

Accuracy and Standard Deviation Data

AnalyteMDLµg/L

Concen-tration*

µg/L

Reagent Water Ground Water Tap Water

Rc SRd R S R R S R

cis-Nonachlor

trans-Nonachlor

Simazine

Toxaphene

Aroclor 1016Aroclor 1221Aroclor 1232Aroclor 1242Aroclor 1248

Aroclor 1254

Aroclor 1260

0.027

0.011

6.8

1.0

0.0815.0

0.480.310.102

0.102

0.189

0.060.450.060.35

25601080

1.0180

3.94.73.63.41.81.72.01.8

110 82 95 86 99 65

– – – – – – – – – – – –

15.2 21.3

9.6 21.8

8.3 3.6

– – – – – – – – – – – –

101 93 83 94 97 59

– – – – – – – – – – – –

7.2 18.3

7.1 17.2

9.2 18.0

– – – – – – – – – – – –

93 87 73 86

102 67

110 114

97 92 86 96

– 84

– 85

– 88

14.3 5.4 4.1 5.1

13.4 6.2 9.5

13.5 7.5 9.6 7.3 7.4

– 9.9

– 11.8

– 19.8

Data corrected for amount detected in blank and represents the mean of five to eight samples.a

MDL = method detection limit in sample in µg/L; calculated by multiplying standard deviation (S) times the students’b

t value appropriate for a 99% confidence level and a standard deviation estimate with n-1 degrees of freedom.R = average percent recovery.c

S = Standard deviation about percent recovery.dR

*Refers to concentration levels used to generate R and SR data for the three types of water matrices, not for MDLdeterminations.–No analyses conducted.

TABLE 3. METHOD ACCURACY AND PRECISION AS FUNCTIONS OF CONCENTRATION-METHOD 505REAGENT WATER

Parameter (µg/L) (µg/L) (µg/L) (µg/L)

Applicable Accuracy as Single Analyst Overall PrecisionConc. Range Recovery X Precision S Sr

Atrazine (3.06-45.90) 1.122C+0.97 0.000 +1.21 0.045 +2.23

Simazine (12.55-50.20) 0.892C+1.446 -0.049 +3.52 0.209 +1.23

Hexachlorobenzene (0.01-0.37) 1.028C-0.00 0.108 +0.00 0.227 +0.00

Lindane (0.04-1.39) 1.009C-0.00 0.057 +0.01 0.142 +0.00

Alachlor (0.50-37.50) 1.004C-0.08 0.077 +0.10 0.105 +0.16

Heptachlor (0.04-1.41) 1.002C+0.02 0.107 +0.01 0.211 +0.02

Aldrin (0.04-1.42) 1.066C+0.00 0.031 +0.02 0.264 -0.00

Heptachlor Epoxide (0.04-1.42) 0.952C+0.00 0.032 +0.02 0.129 +0.02

Dieldrin (0.10-7.53) 1.027C+0.00 0.091 +0.01 0.198 +0.02

Endrin (0.10-7.50) 0.958C+0.01 0.116 +0.01 0.136 +0.02

Methoxychlor (0.20-15.00) 0.950C+0.15 0.115 +0.12 0.125 +0.20

Chlordane (0.51-50.90) 1.037C+0.06 0.084 +0.06 0.125 +0.19

Toxaphene (5.63-70.40) 1.087C+0.24 0.131 -0.31 0.269 +0.69

PCB-1016 (0.50-49.80) 0.856C+0.31 0.106 +0.31 0.147 +0.45

PCB-1254 (0.50-50.40) 0.872C-0.01 0.122 +0.11 0.281 +0.05

*The concentration range applicable to the multi-laboratory study from which the data was generated.

505-25

505-26

505-27

505-28

505-29

505-30

505-31

505-32

505-33

505-34

505-35

505-36

THIS PAGE LEFT BLANK INTENTIONALLY